DocumentCode :
2606532
Title :
Otter: bridging the gap between MATLAB and ScaLAPACK
Author :
Quinn, Michael J. ; Malishevsky, Alexey ; Seelam, Nagajagadeswar
Author_Institution :
Dept. of Comput. Sci., Oregon State Univ., Corvallis, OR, USA
fYear :
1998
fDate :
28-31 Jul 1998
Firstpage :
114
Lastpage :
121
Abstract :
We have developed a compiler, nicknamed “Otter”, which translates ordinary MATLAB scripts into C programs targeting parallel computers supporting ScaLAPACK and several other parallel numerical libraries. We outline the phases of the multi-pass compiler and we discuss the design and implementation of the runtime library. We present the performance achieved by ten compiled benchmark scripts on four distinct parallel platforms, relative to the speed of the MATLAB interpreter executing these scripts on a single CPU. Finally, we illustrate how different data distributions and block sizes can affect the performance of a compiled code
Keywords :
authoring languages; mathematics computing; numerical analysis; parallel programming; parallelising compilers; software libraries; software performance evaluation; C programs; MATLAB scripts; Otter compiler; ScaLAPACK; compiled benchmark scripts; compiled code performance; data block sizes; data distributions; interpreter speed; multi-pass compiler; parallel computers; parallel numerical libraries; parallel platforms; runtime library; Central Processing Unit; Concurrent computing; Design optimization; MATLAB; Parallel architectures; Parallel programming; Personal communication networks; Program processors; Runtime environment; Runtime library;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
High Performance Distributed Computing, 1998. Proceedings. The Seventh International Symposium on
Conference_Location :
Chicago, IL
ISSN :
1082-8907
Print_ISBN :
0-8186-8579-4
Type :
conf
DOI :
10.1109/HPDC.1998.709963
Filename :
709963
Link To Document :
بازگشت